sbi
sbi copied to clipboard
Simulation-based inference toolkit
## What does this implement/fix? Explain your changes This PR implements score-based methods for SBI (and related methods like flow matching). The first goal is to have a running version...
## What does this implement/fix? Explain your changes Implements basic Flow Matchting Posterior Estimation as described by [Dax et al.](https://arxiv.org/abs/2305.17161) ## Does this close any currently open issues? Fixes #963...
There seem to be different local versions of ruff for different contributors, and successive PRs from different contributors change the same files back and forth depending on local versions. Need...
Hi, just a heads-up, if you have scipy version>=1.13.0, you will get an import error because of an issue with arviz. (https://discourse.pymc.io/t/importerror-cannot-import-name-gaussian-from-scipy-signal/14170) In other words, if you do: `from sbi.inference...
As all neural networks in `sbi`, the `MixedDensityEstimator` should also be allowed to have an `embedding_net`. This is especially critical if we intent to use the `MixedDensityEstimator` also to estimate...
Currently, the `build_categoricalmassestimator` does [not take arguments for z-scoring](https://github.com/sbi-dev/sbi/blob/0f0ee6e9f6c35a6d030281427409bcfacc0e4764/sbi/neural_nets/categorial.py#L12-L16). Instead, since it is only actively used in MNLE, `build_mnle` takes care of z-scoring by passing a standarization net to the...
## What does this implement/fix? Explain your changes Implement hNPE, from this paper: https://arxiv.org/abs/2102.06477 This PR adds: - a `hierachical_simulator` decorator to wrap simulator and generate simulation with extra observations....
The [README of the tutorials](https://github.com/sbi-dev/sbi/blob/main/tutorials/README.md) separates tutorials between users and contributors in a too strong way. In fact, all tutorials are for users, not for contributors. In addition, the README...
In #1066, I did not update the `log_prob_iid` method. Instead, MNLE currently uses the standard `log_prob()` to sample [here](https://github.com/sbi-dev/sbi/blob/0f0ee6e9f6c35a6d030281427409bcfacc0e4764/sbi/inference/potentials/likelihood_based_potential.py#L217-L221). For improved sampling speed we should revert to using `log_prob_iid()`
Same as issue #1085 fixed in PR #1086 for the inverse-transform method.